Overview
Independently published modules for the @absolutejs/agent stack. This repository is their source monorepo; npm package names and independent versions remain unchanged.
@absolutejs/agent-modulesworkspacealphaAIIndependently published modules for the AbsoluteJS agent stack.
git clone https://github.com/absolutejs/agent-modules.gitIndependently published modules for the @absolutejs/agent stack. This repository is their source monorepo; npm package names and independent versions remain unchanged.
Workspace — Package — Role
conformance/ — @absolutejs/agent-conformance — Adversarial security and protocol conformance
control/ — @absolutejs/agent-control — Authenticated operator API and console
discovery/ — @absolutejs/agent-discovery — Signed agent discovery
inbox/ — @absolutejs/agent-inbox — Durable verified triggers and delivery
memory/ — @absolutejs/agent-memory — Scoped durable memory
reputation/ — @absolutejs/agent-reputation — Evidence-based reputation
runtime/ — @absolutejs/agent-runtime — Durable agent execution
sandbox/ — @absolutejs/agent-sandbox — Capability sandboxing
trust/ — @absolutejs/agent-trust — Provenance and taint enforcement
Each workspace retains its own license and changelog.
9 public packages and 0 private workspace projects are versioned and developed together.
Outcomes
Independently published modules for the @absolutejs/agent stack. This repository is their source monorepo; npm package names and independent versions remain unchanged.
Workspace — Package — Role
Each workspace retains its own license and changelog.
Hardening checklist
Follow in order
Working example for Development.
bun install
bun run typecheck
bun run test
bun run build9 public packages and 0 private workspace projects are maintained in this repository.
Public package — Provider-neutral adversarial conformance for agent security plus A2A, MCP, Arazzo, and WebMCP standards.
Public package — Authenticated AI agent operator API, console, and bound plan-then-execute playground.
Public package — Signed, searchable, provider-neutral AI agent discovery with live schemas and A2A, MCP, Arazzo, and WebMCP interfaces.
Public package — Durable verified webhooks, event subscriptions, schedules, leases, retries, and dead letters for AI agent triggers.
Public package — Scoped durable provenance-aware memory with authorization, retention, erasure, encryption seams, and pluggable retrieval for AI agents.
Public package — Evidence-based scoped AI agent reputation with W3C Verifiable Credentials 2.0 adapters.
Public package — Durable provider-neutral AI agent runs, steps, checkpoints, budgets, leases, timers, effects, and handoffs.
Public package — Deny-by-default capability grants and provider-neutral HTTP, filesystem, process, and credential-safe action adapters for AI agents.
Public package — Provider-neutral provenance, taint propagation, instruction/data separation, and sink policy enforcement for AI agents.
Scripts declared by this project’s package manifest.